LeoKuo49/whisper-finetune-all

Whisper-finetune_all

This model is a fine-tuned version of openai/whisper-large-v2 on the Amitabha_all dataset. It achieves the following results on the evaluation set:

• Loss: 0.0003
• Cer: 0.2260

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e-05
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer
0.1067	2.5253	1000	0.0800	11.4694
0.0133	5.0505	2000	0.0102	3.3448
0.0017	7.5758	3000	0.0014	0.3232
0.0002	10.1010	4000	0.0003	0.2260

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1